Electronic copy available at: http://ssrn.com/abstract=2390597

DISEI - Università degli Studi di Firenze

Working Paper Series - Economics

DISEI, Università degli Studi di Firenze Via delle Pandette 9, 50127 Firenze, Italia

www.disei.unifi.it The findings, interpretations, and conclusions expressed in the working paper series are those of the authors alone. They do not represent the view of Dipartimento di Scienze per l’Economia e l’Impresa, Università degli Studi di Firenze

Resistance to Change in Recent Italy: Some Elements for a Theoretical Framework

N. Bellanca and S. Innocenti

Working Paper N. 31/2013

Electronic copy available at: http://ssrn.com/abstract=2390597

RESISTANCE TO CHANGE IN RECENT

ITALY:

Some Elements for a Theoretical Framework∗

Nicolo Bellanca1 and Stefania Innocenti (correspondingauthor)2

1Department of Economics and Management, University of Florence, via dellepandette 9, 50127 Florence, Italy. Tel. +39 (0)55 4374565; Fax. +39 (0)55

4374910; email: nicolo.bellanca@unifi.it2United Nations University - MERIT, Keizer Karelplein 19, 6211 TC,

Maastricht, The Netherlands. Tel.+ 31 (0)43 388 4400; Fax: +31 (0)43 3884499; email: innocenti@merit.unu.edu

17 December 2013

Abstract

Italy is driving along a historical glide path. The main reason liesin the fact that in the country, especially in the last three decades,has experienced an extreme form of resistance to change. Everyone,when defecting, prefers that others defect as well, instead of cooperat-ing. This paradoxical situation was denominated Low-Low Game byGambetta and Origgi. In order to clarify how and why the Low-LowGame works, this paper uses parallel games. With the aim of explain-ing the origins, a theoretical framework based on the nexus betweeninvoluntary decisions and loss of power will be developed.

Keywords: Institutional change, Italy, Parallel games, Learnedhelplessness

∗Thanks to Angelo Antoci, Raffaele Bertini, Kaku Attah Damoah, Nicola Doni, DiegoGambetta, Gloria Origgi e Massimiliano Vatiero for their comments and advices. Anyimperfection is ours.

1

1 The Low-Low Game paradox

In a recent paper Gambetta and Origgi (2013) focused their attention

upon a relevant extreme form of resistance to change which seems suitable

to describe the current Italian situation as well as any other more general

cases. The authors shed light on the circumstances wherein everyone, when

defecting, prefers the other players to defect as well, rather than cooperating.

With the aim of presenting the analysis in its simplest way, we consider the

class of simultaneous ordinal 2 x 2 games based on pure strategy. Two players

are considered and assumed to have two distinct strategies each: either they

cooperate (C), either they defect (D). There are, thus, four possible results,

and each of them depends on both players choices. Although the approach

is ordinalist, games are easier to be interpreted through numerical payoffs,

going from 1 to 4. Each player has a ranking of preferences upon the four

possible payoffs. These ordinal preferences are indicated as 4 ≻ 3 ≻ 2 ≻ 1.

In line with the literature, we call R (as “Reward”) the payoff obtained

if both players decide to cooperate (C); P (as “Punishment”) if they both

opt for defection (D); T (as “Temptation”) if player 1 chooses to defect (D)

while player 2 decides to cooperate (C). Ultimately, we call S (as “Sucker”)

the situation in which player 1 opts for cooperation (C), while player 2 for

defection (D).

In the Low-Low Game (Gambetta and Origgi, 2013), represented in Fig-

ure 1, both players converge to the payoff which occupies the second place in

their ranking. The paradox we need to explain regards the fact that P defeats

2

Player 1

Player 2C D

C 2, 2 1, 3D 3, 1 4,4

P≻ T ≻ R ≻ S.

Figure 1: Low-Low Game.

T, which means (D, D) is selected to better off than with the (D, C) strategy:

each one seems therefore to be inclined to renounce to something (i.e. the

cooperation C of the other player) and to obtain a lower reward (i.e. the de-

fection D of the other). We claim that two explanations for this “irrational”

choice can be found. The former consists in reviewing the variables included

in the players payoffs, in order to maintain that each of them reaches a better

final outcome with a lower level of public good. More precisely the “good”

they aim to achieve through cooperative behaviors represents a “damage”.

Secondly the other reason which Gambetta and Origgi (2013) mainly

refer to suggests that agents remain trapped in a “downward” social game

where everyone drowns in the swamp, because each player has internalized

“perverse” social norms based on the reduction of cooperative behaviors.

In our opinion, both these points do not explain all of it. This paper

tries, instead of simply proving the way the problem is displayed, to develop

a theoretical framework which might help in drawing the “unexpected” tra-

jectory the country undertook. We aim at doing this, mainly focusing on

the recent Italian historical facts, but the logic behind this seems suitable

for more general cases. In ¶2 we briefly review the existing literature, in the

third ¶we introduce a discussion based on the notion of parallel games. In

the fourth ¶, instead we propose a mechanism which is generated from loss

of power to leaned helplessness. ¶5 concludes.

3

2 Resistance to Change: background litera-

ture

Habits, customs – the so called informal institutions – as well as laws and

norms have been proved to be exhibiting effects on economic behaviour,

welfare and wellbeing throughout in the course of history all over the planet

(Diamond, 2005; Putnam et al., 1994). Since coined by Kurt Lewin (1946)

the concept of resistance to change has been widely used to describe any

micro, meso or macro process which proves its inability to change overtime.

As a matter of fact this has been a highly debated topic and here, we aim at

scathing a bird eye view of the existing economic and non-economic literature

born over the last 20 years grouping it according to the micro, meso and

macro distinction. No claim of completeness can hold. Moreover, theories

purely devoted to the resistance to change do not exist. Most scholars tend

to reason on the conditions and processes of change to then question the

cases in which these conditions do not hold or seize up, but very rarely some

studies are devoted to the “negative” side of this process as such, and this is

why the above mentioned paper of Gambetta and Origgi (2013) is of great

interest.

Some scholars claim that in principle, resistance to change is a natural

social mechanism. In field theory, it is defined as an equilibrium produced by

the synergy of forces with different intensities and directions (Lewin, 1946).

Being fields generated by individuals’ interaction, change occurs subsequently

to the dismantle of the status quo and follows three steps: from the melt down

of the initial state, to chaos and finally towards a re-solidification of the new

4

state. At any level of analysis (i.e. micro, meso or macro), the notion of

choice lies behind the success or failure of the process of change. According

to Diamond (2005) there exist two characteristics which seem to be crucial

in driving the change towards positive or negatives paths. The first one

relates to the ability of long term planning, the second to the willingness to

revise core values. The courage of acting with a long term strategy implies

anticipatory decisions before the burst of a crisis and constitutes a feature for

successful change. The latter also requires, again at any level, the willingness

to make the due adjustments of core values in spite of their disproportionate

difficulty. A non-dissimilar line of though to explain the success or failure

of development projects is condensed in “The principle of the Hiding Hand”

(Hirschman, 1967). Action at micro and meso level is induced by an error,

being it an understatement of the difficulties of a decision making process

or an exaggeration of its future benefits. Only the lack of awareness of

the efforts required, boosts improvements in social conditions. Moreover,

the effort could not be bypassed without a serious overestimation of the

prospective benefits.

In sociological studies most of the attention has been devoted to the phe-

nomenon of change. Few attention has been given to persistence, the prime

source of predictability and stability without which any society would be

nonexistent (Patterson, 2010). According to Patterson (2010) instead, tak-

ing institutional dynamics as a given, most of the attention should be focused

on the “persistence puzzle”. Some scholars as Bourdieu and Nice (2000), al-

though acknowledging that change is an incremental process driven by errors

or deliberate choices, claim that the notion of habitus explains both the

5

cultural and, indirectly also, the structural reproduction. According to Pat-

terson instead, a unique double-edge concept is not suitable and he claims

that institutional reproduction is led by 7 distinct but combinatorial mech-

anisms. Social learning and imitation allows inter and intra-generational re-

production. Secondly, reproduction occurs if some rules of the game become

normative and part of the shared definition of reality. A third way for repro-

duction to become explicit is through the channel of hegemony or through

culture specific cognitive processes. A forth mechanisms explaining repro-

duction is frequency driven selection: individuals excessively select a specific

declination of the cultural process on the basis of its popularity. Reproduc-

tion is strictly related also to the communication dynamics in a community

which produce specific individual interactions. The basic idea is that com-

munication aims at transforming shared knowledge to a common ground of

understanding minimizing confusion and uncertainty. Re-interpretation is

instead a hidden form of persistence and consists in portraying a process and

its meaning or practice in terms of others. Lastly, reproduction is originated

by “embedded introjection”. By transforming a clear pattern of beliefs into

a spiritual one placed at the core of the dominant cultural institution, a

collective imaginary is created and it will stay dormant for possibly long.

A useful explanation, straddling the macro, meso and micro classifica-

tion, is the one proposed by Banerjee and Duflo (2011). Resistance, and

therefore unsuccessful process of change, results from “three Is”: firstly from

the absence of adjustments (Inertia); secondly from the claim to have the

best solution, without assuming any alternatives (Ideology), and lastly from

uncertainty over future outcomes (Ignorance).

6

The most important neo-institutional strand of literature considering

meso and macro phenomena is the one that goes from Robert Bates with

Prosperity and Violence (2001), to the Institutions-Organizations-Beliefs frame-

work proposed by North, Wallis and Weingast (2009), up to the recent work

of Besley and Perssons (2011; 2010) and Acemoglu and Robinson (2012).

Bates argue that economic development and the transformation of violence

into a publicly managed good are two parallel and mutually reinforcing pro-

cesses. Nesting his reasoning on Tilly’s work on early modern state formation

(1975), Bates (2001) argues that pre-modern societies are characterized by

effective but inefficient private provision of violence. Once this inefficiency

is acknowledged, central authorities and decentralized interest groups get

organized. This, together with industrialization, paves the way to public

provision of violence. Violence changes therefore its nature: it is not a mean

of predation anymore but rather a productive resource for protection and

capital accumulation. The “domestication of violence” is hence the motor of

successful change. Along these lines, North et. al. (2009) claim that politics,

and particularly coercive power, deeply affects countries’ opulence and secu-

rity as well as the ranges within which they vary. Intra-elite coalitions and

repartition of economic rents, the adaptive capacity of institutions and the

nature of informal institutions are key determinants of growth and volatil-

ity. The different ways in which power is used, determines the ability of

societies to face shocks either collapsing either adapting. Societies wherein

citizens can openly access any economic and political organizations, expe-

rience positive and stable growth paths. Closed-access societies, in which

the distribution of economic returns and access to organizations is heavily

7

skewed towards elites, are characterized by volatile growth, scarce democratic

consent, highly centralized governments, personal relationships, high levels

of inequality. Besley and Perssons (2011; 2010) claim that credible and co-

hesive political institutions are of crucial importance in generating change.

Political competition is the determinant as to how cohesive the institutions

will be chosen to be. If elites feel no threat exists, they are less likely to

create systems of control than if the possible loss of power is feared. This

strongly recalls Acemoglu and Robinson’s theory who argue that inclusive

political and economic institutions, are the main drivers of economic devel-

opment. Inclusive political institutions are defined as sufficiently centralized

and pluralistic political structures in which absolute power cannot appear.

Conversely, the presence of extractive institutions, which prevent widespread

participation and discourage productive activities, is the main reason for un-

successful processes of change. It this therefore the set of interests held by the

political power which might generate institutional persistence. The lack of

binding commitment, which intrinsically affects every political arrangement,

allows for predatory behaviors of the ruling party leading to inefficiencies

(Acemoglu, 2003).

The meso explanation of resistance to change par excellence is the one of

groups’ particularism at the core of Olson’s work (1965). A group is made of

agents holding the same interests, but contemporary every group has its own

interest. Single individuals’ goal would not be achieved without collective

action. The idea that groups act supporting their group interest follows

from the assumption of rational and self interest behavior. However, this is

not completely true. In fact, individual rationality in seeking higher level

8

of personal welfare does not match the group objective to advance towards

a common goal. This only occurs in presence of either a coercive force or

separate incentives given to group members to work for a common intention.

The lack of those elements at meso level implies the reduction of any more

macro process of change. According to Putnam et al. (1994) instead the

lack of change can be traced back to the scarce density if social capital. The

denser are the networks of civil engagement – i.e. the quintessential form

of social capital – the higher is the ability for individuals to cooperate for

mutual benefits and therefore the more this should foster institutional success

in the broader community.

Within the neo-insitutional strand of literature, David (1985; 1994) and

Arthur (1989; 1994) carry out a micro analysis. Their notion of path-

dependence, is meant to capture the way in which small, historical contin-

gent events can generate self-reinforcing mechanisms which might lock the

economy into not necessarily efficient particular structures and pathways of

development. Resistance to change hence is the fruit of increasing returns,

self-reinforcement, positive feedbacks and lock-in. The presence of increas-

ing returns implies that the more frequently a decision is adopted the higher

the benefits it produces (Pierson, 2000). Self-reinforcing mechanisms instead

imply that carrying out a choice activates a set of forces or institutional

complementarities which encourage the support of the first decision made.

In case of positive feedbacks, once a decision is made it might generate pos-

itive externalities as soon as the same action is put in place by other agents.

Lock-in instead means that some choices or actions become more preferable

than others because of the high number of people following them.

9

But the true micro-level process which produce resistance to change can

be traced back to conflict theory of decision making which intrinsically re-

calls the cognitive structure of agents. The main concern of normative and

descriptive decision theory is to understand how a decision maker is will-

ing to act in a given choice setting. The concept of willingness involves a

series of values according to which a decision is said “good” or “bad” (i.e.

a decision that is regretted to have been taken). Nevertheless, sometimes

conflicting situations arise and constant and systematic analysis of the alter-

natives is impossible. In these cases a bolstering process starts (Janis and

Mann, 1977). Bolstering implies the agent, when evaluating all alternative,

will filter relevant information with the purpose of supporting his preferred

option. All information which sustains rival alternatives is suppressed. This

tendency shows a preference for avoiding rather than solving conflictual sit-

uations. The bolstering phenomenon is analyzed in psychological literature

under the name of cognitive dissonance. Cognitive dissonance according to

Festinger (1962) refers to a situation involving conflicting attitudes, beliefs or

behaviors which generate a sense of discomfort and dissatisfaction in the in-

dividual. This feeling is alleviated when one alternative wins over the others

or all of them are made compatible. While cognitive dissonance expresses

itself in the post-decisional phase, bolstering occurs before the decision is

made binding. The distinction between the post-decisional rationalization –

at the core of cognitive dissonance theoretical framework – and pre-decisional

bolstering – the center of the conflict decision theory – seem to threaten the

freedom of choice of individuals producing compliance or defiance which, on

their hand, enable reaction to the social milieu (Rapoport, 1988). This is

10

the aspect this paper focuses upon trying to shed some light on the Italian

glide path using parallel games.

3 An explanation based on parallel games

Every human being, as a member of a collectivity, is constantly involved

in situations of strategic interaction with others whose interests might be, at

least partially, in contrast to hers/his. These multiple situations concern (at

least) the market, bureaucracy, democracy, job placement, political activities,

familiar relationships and friendships, as well as national, ethnic and religious

identities. The hypothesis we start from suggests that, very frequently, the

various strategic interactions are connected with one another. What occurs in

the market shapes what happens in the political sphere; national belonging

affects religious choices; affective life influences working outcomes, and so

on. It follows that a joint analysis of the occurrence of many situations can

explain phenomena that would not be captured if the analysis was conducted

separately. When an individual is engaged in several strategic interacting

situations, either s/he can meet different agents in various occasions, either

s/he relates to the same individuals in some others. In the first case, in line

with Alt and Eichengreen (1989), we refer to overlapping games, while in the

second case we deal with parallel games. We hereby focus on parallel games

only, as in our opinion, they are particularly effective in revealing how and

why the relationship among certain subjects is implemented only in case of

multiple strategic situations. The 2 x 2 simultaneous ordinal games based

on pure strategy are 726 (Kilgour and Fraser, 1988). Only 78 of them are

11

ordinally different, meaning that all payoffs vary (Rapoport and Guyer, 1967;

Robinson and Goforth, 2005). For the sake of our argument we chose five

games, whose interpretation appears able to represent recurring and relevant

situations. The first, yet introduced, is the Low-Low Game (LLG). The

second is the No Conflict Game (NCG) instead. As represented in Figure 2,

in this case both players chose the payoff they placed first in their rankings.

It depicts a social situation where the optimal scenario is reached without

any contrasts, as both players share the same preference. Player 1 obtains 4

as payoff if R; he obtains 3 if T; he obtains 2 if S; lastly, he obtains 1 if P.

The same applies for player 2.

Player 1

Player 2C D

C 4,4 2, 3D 3, 2 1, 1

R≻ T ≻ S ≻ P

Figure 2: No Conflict Game.

In the Second Best Game (SBG), represented in Figure 3, both play-

ers chose the payoff placed second in their rankings. It is a social situation

where, even without reaching the optimal scenario, players converge to rela-

tively high payoffs. We can interpret this game by affirming that, although

excluding perfection, the Second Best constitutes the most realistic and sat-

isfactory situation. Player 1 obtains 4as payoff if T; obtains 3 if P; obtains

2 if S; lastly, obtains 1 if R. Player 2 obtains payoff 4 if R; obtains 3 if P;

obtains 2 if T; lastly, obtains 1 if S.

The well-known Prisoner’s Dilemma Game is here renamed Third Best

Game (TBG), as one of its characteristics, defining its difference from the

12

Player 1

Player 2C D

C 1, 4 2, 2D 4, 1 3,3

Player 1: T≻ P ≻ S ≻ R

Player 2: R≻ P ≻ T ≻ S

Figure 3: Second Best Game.

SBG, regards the fact that both players chose the payoff placed third in their

rankings. It is a social game, represented in Figure 4, where free-riding – i.e.

an uncaring attitude – triumphs and the optimal scenario is far from being

reached. Player 1 obtains as 4 payoff if T; obtains 3 if R; obtains 2 if P;

obtains 1 if S. The same applies for player 2.

Player 1

Player 2C D

C 3, 3 1, 4D 4, 1 2,2

T≻ R ≻ P ≻ S

Figure 4: Third Best Game.

Lastly, in the Hegemony Game (HG), represented in Figure 5, the domi-

nant strategy leads player 2 to voluntarily choose the payoff placed third in

her/his ranking, whereas player 1 obtains the payoff placed first. A possi-

ble interpretation might be that player 2 is exploited or expropriated by the

other agent because s/he accepts a clearly asymmetrical status quo. Player 1

obtains payoff 4 if T; obtains 3 if R; obtains 2 if S; and obtains 1 if P. Player

2 instead obtains payoff 4 if R; he obtains 3 if T, 2 if S; and 1 if P.

13

Player 1

Player 2C D

C 3, 4 2, 3D 4,2 1, 1

Player 1: T≻ R ≻ S ≻ P.

Player 2: R≻ T ≻ S ≻ P

Figure 5: Hegemonic Game.

Our idea is that a reason for selecting the LLG, interpreted as a form of

resistance to change, resides in the fact that some games are carried out in

parallel. Playing an LLG becomes rational as long as it is conducted together

with other specific games. With no claim of elaborating an exhaustive anal-

ysis, we suggest some parallel games which contribute to clarify the concrete

historical phenomena amongst which the one Gambetta and Origgi refer to,

stands out: the Italian socio-economic decline. The latter can be traced back

to the incapacity of the country to open up to change.

We call the first type of parallel games of “the Count and the Baroness”.

It involves both No Conflict Games and Low-Low Games. We introduce it

through a curious and apparently far-from-relevant question: why, in the

Courts of pre-modern Europe, did noble males rigidly stick to honor codes

based on the exclusive possession of their wife, while at the same time hav-

ing libertine behaviors which led everyone to flirt with others’ wives? Let

us assume that the Count is bored of his Countess and wishes to flirt with

the Baroness. He would be glad if, while accomplishing his goal, the Baron

did not intercourse with his wife otherwise he would violate the honor code,

while the Baron would cooperate with it. However, the Count is aware that

the honor code is actually implemented in the Court perimeter and that, as

14

soon as the Baron finds out about the flirt, he would kill him. Hence the

Count is better off when playing within two distinct but interacting situa-

tions, that we will call “tables” from now on. On the formal rule table, all

agents act, without disagreement, according to the honor code and declare to

be jealous: the Count flirts with the Countess and the Baron flirts with the

Baroness, namely they all cooperate (NCG). Contrarily, on the silent agree-

ment table, an exchange of spouses occurs: the Count enjoys the Baroness’s

company just because he knows the Baron is doing the same with his wife,

the Countess, meaning that everyone defects (LLG). Figure 6 matches the

two situations within the same game, whose payoffs are the sum of the payoff

obtained in the separate games. The equilibrium position (8,8) results from

the dominant strategies of the two players.

Player 1

Player 2D,D D,C C,D C,C

D,D 5, 5 7, 6 4, 2 6, 3D,C 6, 7 8,8 5, 4 7, 5C,D 2, 4 4, 5 3, 3 5, 4C,C 3, 6 5, 7 4, 5 6, 6

Figure 6: Joint Game

As we have claimed in ¶1, it would be irrational to play the LLG sepa-

rately. However, the (D,D) strategy of the LLG is able to justify and auto-

reproduce itself if it is jointly enforced with the (C,C) strategy in the NCG:

every husband de facto flirts with everyone else’s wife as officially every wife

is faithful to her own husband. The theoretical point is that we can explain

the priority of the defecting strategy on the second table as it prevails over the

15

cooperative strategy performed in the first table, and the other way around.

The parallel games are able to make this reasoning mode explicit. Although

the situation of the noble Court can be perceived as far and factitious, its

logic constantly reoccurs in the Italian case. The systematic discrepancy

between formal rules and silent agreements is crucial when aiming at under-

standing economic activities which elude and evade taxes, which stipulate

fake contracts or circumvent norms (D’Agnese and Abravanel, 2011). The

game of the Count and the Baroness explains that the gap endures just be-

cause the agents simultaneously play on both tables. As long as we focus on

one table at the time, we can describe the discrepancy between formal and

silent rules, but we do not explain it.

Let’s move to the second type of parallel games, which we call of “the

Neapolitan traffic”. Drivers in Naples often violate the rules of the formal

driving code: traffic lights are ignored, cars are parked wherever a free spot is

available, motorcyclists do not wear the helmet, car drivers do not wear safety

belts, and so on. On this first table a LLG is performed. However, we would

hardly understand the functioning of this urban system without consider-

ing the existence, at least in some cases, of a second informal table wherein

players opt for a SBG. For instance, as traffic lights are not considered, the

traffic flow is never stuck, it rather keeps going at a lower but constant speed.

This method, if adequately shared by all players, can generate a fluid traffic

flow which is even more regular than if traffic lights were respected. First

of all, it has been proven that driving at the same speed of the traffic flow is

safer (Lave, 1985). Secondly, adapting to the other cars speed makes the risk

of being sanctioned lower, as the police can only catch a small proportion

16

of traffic rules-breakers (Dixit and Nalebuff, 2008). Lastly, cars stopping or

moving according to traffic lights do not guarantee the avoidance of traffic

jams.

Player 1

Player 2D,D D,C C,D C,C

D,D 7,7 8, 5 6, 4 7, 2D,C 6, 6 5, 8 5, 3 4, 5C,D 4, 6 5, 4 5, 5 6, 3C,C 3, 5 2, 7 4, 4 3, 6

Figure 7: Joint Game.

Figure 7 considers the two situations in a single game, whose payoffs are

the sum of the payoffs obtained by means of separate games. On the first

table the two players are engaged in a LLG: they obtain (4,4) as a payoff.

But, as said, it would be irrational to play the LLG separately. On the second

table they challenge each other performing a SBG and they both obtain 3

as a payoff. The SBG generates rational incentives for both players and can

be played autonomously. If an agent only plays a SBG s/he obtains 3/4=

75% of total possible payoffs. Everyone has hence the incentive to jointly

perform a LLG and a SBG as s/he gains the 87% of available payoffs. In

other words, the (D, D) strategy of the LLG, although it cannot justify itself

autonomously, it is able to auto-reproduce itself if jointly enforced with the

(D,D) strategy of the SBG. The counter-intuitive downside is that, if the two

games are played jointly, everybody improves his condition with respect to

the case in which only the SBG is played. Everybody obtains higher welfare

coupling the SBG –“realistically more efficient”– with a “downward” game

17

– hence highly inefficient – such as the LLG.

We call the third type of game of “ the plagiarist Economist”, wherein

a SBG and a LLG simultaneously occur (as in the above mentioned case)

but here a HG adds up. In their paper, Gambetta and Origgi (2013) argue

upon a case they reckon emblematic of the Italian decline: the plagiarism

carried out by one of the most famous Italian scholar in economics. On the

first table the Economist (B) provides the Academia (A) with an essay of

adequate quality: a Second Best Game is performed. On the second table

player B copies the essay while player A pretends not to be aware of it: a

LLG is performed. However, the most relevant point in our opinion, which

Gambetta and Origgi do not focus explicitly upon, concerns the existence of

a third table. It is not B the one who plagiarize, but rather a player C, a con-

tract research for instance. In order to interact with A, player B has to exert

power on C, following the logic of the hegemonic game. The researcher who

writes for B is socially and institutionally invisible: we can indirectly discover

his presence only when, as it actually happened in the episode recalled, he

changes his strategy and decides to defect without being concerned about the

simple cautions that prevent plagiarism from being revealed. Thus, while on

the third table only A and B seem to be still playing, C secretly takes part

in the game and can sometimes overthrow the whole result of the parallel

games. As we can easily notice, the logic of the plagiarist economist captures

other relevant aspects of the Italian decline. Lets analyze another example:

in Italy, an Entrepreneur (E) offers a temporary contract to a recently grad-

uate student (G). On the formal rule table, G has full autonomy and flexible

working schedule and earns a net income of 600 euro per month. On the

18

silent agreement table both players know that G will work in the companys

office, he will have E as supervisor and a fixed working time plan of at least

40 hours per week. How is G selected? G’s place could be taken by whoever

is willing to play on both tables, which means that the job goes to who is able

to be E’s accomplice. In turn, the job is only taken by who plays with E on

a third table, on which another “shadow player” (F) – G’s family – operates,

paying the difference between the performances declared on the first table

and those carried out on the second. In brief, the players are only G and E

(and the payoffs added up are theirs), but on the third table G is supported

by F; both G and E know about the existence of F, but they play on the

first two tables as if F did not exist. We are assuming that the three games

(SBG, LLG, HG) correspond to an equal number of institutional tables, on

which the same players simultaneously perform. On the first table players

operate according to formal, official or explicit rules. On the second table

instead, silent agreements or informal rules apply. Lastly, the third table

constitutes the underlying, or structural, social situation which fosters the

conditions thanks to which the other tables are able to function. Therefore,

the explanation is partially different from the one concerning the noble Court

or the Neapolitan traffic: in those cases, it was sufficient to show that the two

tables were justifying each other; here, instead, there is an additional table

without which the two others would not work. Whereas in the preceding

types of parallel games we have assumed the players being equal, both with

respect to the social status and the de facto power; on the contrary, when

considering also the HG, the two players are assumed to be asymmetric, in

terms of power and resource endowment. Let’s now go through the trend of

19

payoffs in this type of parallel games. On the first table the two players are

engaged in a SBG: they both obtain 3 as payoff. On the second table they

challenge each other with a LLG: they gain (4,4) as a payoff. Lastly, on the

third table they face a HG, where one player gains 4 and the other obtains

2. What matters, for each of them, is the sum of the three payoffs. If player

1 only carries out a SBG he obtains 3/4 = 75% of the total available payoffs.

If s/he performs a SBG and a LLG he gets 3/4 + 4/4 = 7/8 = 87% of the

total available payoffs. If s/he performs a SBG, a LLG, and a HG instead

he reaches 3/4 + 4/4 + 4/4 = 11/12 = 92% of the total available payoffs.

Hence, player 1 is better off if s/he performs jointly in the three games. Also

player 2 gains, as the other player, the 75% when performing only a SBG,

and the 87% when performing jointly a SBG and LLG. Contrarily, if s/he

contemporary carries out the 3 games, player 2 will only obtain 3/4 + 4/4

+ 2/4 = 9/12 = 75% of the total available payoffs. Hence player 2 worsens

her/his position if performing jointly in the three games. The logic is hence

made clear: as long as we consider two tables as it happens in the Neapolitan

traffic game everyone achieves the best outcome by choosing a LLG and a

“reasonably efficient” game such as a SBG. However, when introducing a

third table characterized by asymmetry, then the spell is broken and the less

powerful agent additionally worsens her/his position. Lastly, let’s assume

that on the third table the game form changes, and turns from HG to TBG.

This means, considering for example the plagiarist economist case, that the

contract researcher is no longer dominated by the economist; s/he stops col-

laborating and starts defecting. Instead of “copying in a collaborative way”,

the young contract researcher will literally translate the pages written by an

20

internationally known author, plugging them in the paper that the famous

Italian economist will sign. The plagiarism is then revealed. Now both player

1 and 2 perform a SBG, a LLG, and a TBG obtaining 3/4 + 4/4 + 2/4 =

9/12 = 75% of the total available payoffs. Hence, both players worsen their

positions when performing jointly the SBG, a LLG, and a TBG. This implies

that the incentive to jointly play in the three institutional tables vanishes.

Concluding, parallel games represent a useful conceptual tool to analyze

interactions between miscellaneous social situations. Such tool helps in ex-

plaining in several ways the rationale for the selection of the LLG. We focused

mostly on three important cases to understand the peculiar Italian situation:

the coexistence of a NCG and of a LLG as in the case called of the Count

and the Baroness ; of a LLG and a SBG as in the case denominated of the

Neapolitan traffic; of a SBG, a LLG and a HG (or, alternatively, of a TBG)

as in the case named of the plagiarist economist.

4 A mechanism generating the Low-Low

Game

A close examination of parallel games contributed to clarify, with a par-

ticular reference to the current Italian situation, how the Low-Low Game

can function and why it might be voluntarily chosen. In this paragraph we

will try to identify an explanatory mechanism – not necessarily the only one

– that might give origin to the Low-Low Game. The initial move of our ar-

gumentation consists in taking into consideration two categories of economic

21

goods that, differently from usual private goods, are both based, although

with different signs, on joint consumption1. The first category is that of re-

lational goods (RG): a subject’s positive consumption grows if other people

increase their positive consumption of that same good. A RG cannot be

consumed by a single individual, because it relies on interactions with other

subjects and can be used only if shared. Watching a soccer game in a sta-

dium represents, among others, a common example: the higher the number

of supporters of the team the agent stands for, the higher is her/his own

positive consumption. The other category is that of positional goods (PG):

agents’ positive consumption increases if others consume an increasingly neg-

ative amount of that same good. Examples of crucial importance are prestige

or power goods: it is possible for a person to benefit from them only at the

expenses of others. If a dominant subject exists, someone who is dominated

exists as well; if there is a person with a higher social status, someone with

a lower social status therefore exists. If we recall a classic distinction by

Leibenstein (1950), RG produces a bandwagon effect: the higher is the gen-

eral consumption of a good, the higher the incentive to consume it. Hence,

the RG fosters forms of conformist behavior. On the contrary, PG produces

a snob effect: the lower is the general consumption, the higher the push to

take advantage of it. The PG, thus, favors opposite forms of behavior with

respect to RG.

Let us imagine a society divided into two types of groups. Group type C

holds “conformist” behavior and seeks for uniformity of consumption choices,

1Here we closely follow Fiorito and Vatiero (2013). For Relational goods confrontUhlaner (1989), whereas for Positional goods confront Pagano (1999).

22

whereas group type P holds “positional” behaviors and seeks for individual

relative advantage, meaning that every member of the group aims at differ-

entiating her/himself from other members. In a society there can be several

C groups aiming at reaching an internal homogenization, but maintaining

differences with respect to other groups. In recent Italy, for instance, profes-

sional corporations such as those of notaries, pharmacists, taxi drivers, estab-

lished rules to create strong internal behavioral regularities for the members,

but at the same time maintained the due differences to separate them from

other corporations. On the other hand, there are groups P which, aim both

at differentiating one another and promoting an inner positional competition.

As Thomas Marshall (Marshall, 1964, p.198) states:

Social status rests on collective judgment, or rather a consensus of

opinion within a group. No one person can by himself confer status

on another, and if a man’s social position were assessed differently by

everybody he met, he would have no social status at all.

Boats’ owners, for instance, are very different from the members of book

clubs; within each group a specific competition to have the biggest boat or

to become a member of the most prestigious club applies. Let us assume

that both groups, C and P, can choose between two strategies as represented

in figure 8. The first strategy is denominated H and consists in aiming to

high quality results; the other is instead named L and aims to get to low

quality outcomes. For those aiming at a RG, a result H consists in the

occurrence of a trickle-down effect: what is achieved by a limited number of

people is also progressively obtained by the other members of the group, thus

enhancing homogeneity and conformism. When group C chooses strategy

23

H, a “higher” egalitarian tendency arises. Considering private goods, this

is the case of middle classes striving to get the consumption levels of the

ruling classes, while lower classes strive for the level of consumption of middle

classes. As it happened in Italy during the years of the so-called “economic

miracle”, a social mobility elevator is triggered: the director purchases a

car, a washing-machine or an apartment but it will take five and ten years

respectively for a manager or for an executive worker to purchase the same

items. Although everyone strives for the same good, they cannot have it

in the same amount of time. What happens for private goods, also occurs

for RG. The latter are a class of local public goods, for which non-rivalness

and non-excludability matter only within a certain set of relations (Uhlaner,

1989). As a consequence, RG can be used by a single individual jointly with

other members of the social network. During the Italian “economic miracle”,

the set of local public goods – from education to the health care system, from

social security to solidarity, and so on – got institutionalized and expanded,

contributing to an upward social mobility. This is represented by cell CH in

Figure 8.

As positions are socially scarce, for those who decide to have a PG, H

is the result of a winners-take-all competition (Rosen, 1981). Indeed, in or-

der to distinguish Tom from James, the positional good must create social

scarcity, through exclusion mechanisms based on pricing or limited access. If

everyone could afford or access a tropical beach, there would no longer be a

differentiation between agents. Given the “social scarcity” of positions, only

someone wins, while many others lose. The higher the number of groups type

P which choose high quality outcomes (H), the wider the spectrum of pur-

24

sued PG. However, if a large number of groups P purses an equal number of

PG, the society tends to diverge on multiple scales of values and goals. This

might also imply a blooming of innovative actions, as innovation is fostered

through the refusal of conformism and love for variety. This is represented

by cell PH of Figure 8.

H L

C “Upward” Egalitarianism “Downward” EgalitarianismP Competition for differentiation Defense of gained positions

Figure 8

When a group C selects strategy L, a form of “downward” egalitarianism

occurs. As it happens in today’s Italy, the social mobility elevator goes

down for (almost) everyone (Pianta, 2012). Along this glide path, the role

of RG is crucial: indeed, the set of local public goods has gradually been

flaking in Italy throughout the last three decades, with increasingly serious

consequences which shifted down the social scale even more (ISTAT, 2013).

This is represented by cell CL in Figure 8. Lastly, when a group P opts for

strategy L, the positional internal competition is blocked or weakened. This

means the members of that group would fight for a specific good which confers

them a specific status. This good can be obtained by few at the expense of

many, but these few are institutionally able to compete. The differentiating

(and sometimes innovative) dynamics of positionalities has ceased to apply.

A widely discussed example in Gambetta and Origgi (2013) is represented

by the functioning of the university system in Italy. In a context that should

be promoting a “virtuous” positional competition, an institutional arrest

25

is observed. This is represented by cell PL of Figure 8. Hence, column

H represents a dynamic society. Cell CH describes an inclusive dynamics,

whereas PH portrays a dynamics based on distinction. In a dynamic society,

the “upward” egalitarianism of cell CH and the winner-take-all competition

of cell PH coexist. While CH includes wider social groups, PH differentiates

the elites (being or self-representing them as such). Contrarily, column L

represents a declining society. Cell CL describes an inclusive decline, while

cell PL pictures a privileged downfall. In a declining society, the “downward”

egalitarianism of cell CL and the weakening of positional fight represented

by cell PL, coexist. CL excludes wider social groups, whereas PL tends to

normalize elites, transforming them into groups that, rather than gaining a

new status, firmly hold what they previously achieved.

How is it possible for a society to shift from column H to column L?

In order to answer this question it is at first necessary to state that such

transition entails power loss processes, herewith called disempowerment. In-

deed, the members of the several groups – either pursuing RG or PG – feel

that the spectrum of their actions has shrunk: while in column H upward

mobility and competition for distinction is possible, in column L there are

possibilities for downward mobility and leverage of status goods. This pa-

per does not aim at examining the historical concrete circumstances that

fostered the intra-national or international disempowerment of groups of the

Italian society (this analysis has been carried out especially by historians:

see, for example, Ginsborg (1998)). In general, the goal is to suggest the

hypothesis according to which an individual tends to interpret a condition in

terms of disempowerment when, in that situation, s/he cannot take voluntary

26

decisions.

Involuntary economic decisions are at the core of Trygve Haavelmo clas-

sical essay of 1950. A wide part of the economic theory, writes Haavelmo,

states that any action is rational, as long as an individual prefers it to any al-

ternative and as long as the market allows its fulfillment. It follows that any

market equilibrium is a voluntary situation, as each agent tends to choose the

best option within a set of circumstances which make her/his actions com-

patible with those of the others. Haavelmo observes that, in order to define a

decision “involuntary”, two conditions should be in place. Firstly, the agent

has to be able to compare the current situation to an alternative practicable

one; secondly, the agent should be able to realize that the alternative option

is outside his/her personal scope of action, as it requires a collective action

and therefore is not feasible. In this precise intersection

individuals can be aware that some market equilibria are preferable

to the existent ones, but if the necessary actions to achieve them are

out of their scope of opportunity, they cannot do anything else but

choosing the desired actions within the undesired context they are

in. This entails the existence of macroeconomic facts which cannot

be considered a simple sum of microeconomic decisions: institutional

mechanisms, such as the market mechanism itself, as well as social

conventions or opinion changes, limit individual action without be-

ing influenced other than in an indirect and obscure way (Dardi and

Targetti, 1984, p.87,authors’ translation).

Haavelmo’s important definition does not concern only economic decisions.

The rise of involuntary decisions lays the ground for a possible transition from

27

column H to L: Agents are often forced to make choices that do not want

to, as alternative voluntary decisions lay beyond her/his sphere of interven-

tion. Hence, the inter-subjective condition that makes involuntary decisions

possible constitutes the premise for the subjective condition of disempow-

erment. We are now at the second step of the reasoning: a loss of power

(disempowerment) becomes impotence (powerlessness) when the individual

faces processes that s/he perceives as uncontrollable. While disempowerment

arises when the subject is not able to edit an event, powerlessness depicts

any situation in which the agent is absorbed by an uncertain radical event

which therefore falls outside her/his control. These two experiences might

be contemporary but distinct. In case of disempowerment the agents has the

awareness that the set of accessible alternatives for which is possible to cal-

culate the risk has shrunk. In case of powerlessness the agent faces forms of

uncertainty about the nature of the alternatives, the links between them, as

well as about the outcomes that will arise from those links. A loss of power

that slips away from the agent’s control, and not only from her/his ability of

intervention, is perceived as powerlessness. According to Rucker and Galin-

sky (2008), the most immediate reaction to powerlessness is an increase of

attention towards PG, as they are symbols of power.

Power relates to one’s relative control over valued resources, whereas

status relates to the respect one has in the eyes of others. Despite these

conceptual differences, evidence suggests that power and status can

compensate and substitute for each other. [. . . ] When one’s power

is threatened, individuals appear to place a greater value on products

explicitly linked to status, think about products in terms of the sta-

28

tus they convey, and place greater value on monetary wealth (Dubois

et al., 2012, p.1049).

The implication is that in a society wherein most groups loose power, the

immediate response is often the pursuit of status and PG. This theoretical

framework effectively applies to the situation Italy experienced over the last

three decades, when an atomization of the communitarian fabric, individu-

alistic ambition for success, exasperated exhibitionism and on the political

ground the triumph of “Craxism and Belusconism” occurred (Dei, 2011).

We finally reached the third and last step of our reasoning. In Italy,

many people have strengthened propensity towards PG right when the en-

tire country was moving from column H to L, and in particular from PH

to PL. In the winners-take-all competition, typical of the cell PH, the ma-

jority of players lose power, while very few win. This generates temporary

forms of disappointment and frustration when the game appears as fair and

repeatable to whom is involved into it. Those who lose today still hope to

win next time. The most common example is the continuous creation and

implementation of national lotteries: almost no one wins, almost everyone

loses money, but no one complains as everyone perceives the game as fair

and repeatable. Contrarily, when agents, as in the situation described by the

cell PL, strive for a status and a PG, but the winners-take-all competition

is stuck or weakened, disappointment and frustration endure for long time.

This, in turn, additionally strengthens and diffuses the initial perception of

powerlessness to which agents had tried to react through the pursuit of a

PG. The ultimate result is that voluntary decisions tend to vanish from the

conceivable possibility set and, even when concrete opportunities emerge, the

29

individual will not catch the possibility to change the status quo. This be-

havior is denominated learned helplessness (Abramson et al., 1978; Martinko

and Gardner, 1982). Figure 9 summarizes this framework.

9.jpg

Figure 9: The Framework

Hence, when all attempts in regaining power repeatedly fail, powerlessness

turns into learned helplessness. The latter induces a reduction of self-efficacy,

namely of cognitive construction of actions. The pejorative beliefs on self-

efficacy lower agency, that is the level of control upon our own actions, as

well as the pursuit (and acknowledgment) of personal desires. The presence

30

of a significant number of subjects affected by learned helplessness within a

group generates inertia in the group itself. Inner interactions are reduced and

collective passive behaviors emerge: Tom feels like he is justified, for instance,

to silently accept the presence of the members of the Camorra in Scampia

an area of Naples, if James and Harry will do the same. The reduction

of self-confidence as a consequence of learned helplessness, can occur also

simply due to group membership where mechanisms of contagion based on

mimetic attitudes emerge. If James and Harry do not act even when it

would be possible to do so, Tom consequently will tend not to contrast their

behaviors. As result any PG is waived, even when there is a concrete occasion

to compete for it. Agents get involved in a “downward” game in a conformist

manner. They move from cell PL to CL. Obviously, even in today’s wrecked

Italy there are social groups who continue playing in the other three boxes

of Figure 9. But it is cell CL, and the Low-Low Game it represents, that

better describes the actual condition of the country.

5 Concluding Remarks

In this paper we have introduced and discussed some theoretical elements

– parallel games, the nexus between involuntary decisions and loss of power –

that might contribute in explaining the extreme cases of resistance to change

occurring in today’s Italy. The entire paper revolves around the paradoxical

logic of Low-Low Game, according to which everyone when defecting prefers

other players to defect as well, instead of cooperating. The paradox is that,

as displayed in Figure 1, (D,D) confers higher payoffs (4,4) to both players. It

31

follows that players gradually slip off in “quicksand”, hanging each other on

a mutual connivance that makes any alternative worse at their eyes. Feeling

fine, while entering a vicious spiral, explains the reasons for both the absence

of any vital reaction and the passive perpetuation of the Italian glide path.

References

Abramson, L., M. Seligman, and J. Teasdale (1978). Learned helplessness

in humans: critique and reformulation. Journal of abnormal psychol-

ogy 87 (1), 49.

Acemoglu, D. (2003). Why not a political coase theorem? social conflict,

commitment, and politics. Journal of comparative economics 31 (4), 620–

652.

Acemoglu, D. and J. Robinson (2012). Why nations fail: the origins of power,

prosperity, and poverty. Random House Digital, Inc.

Alt, J. E. and B. Eichengreen (1989). Parallel and overlapping games: Theory

and an application to the european gas trade*. Economics & Politics 1 (2),

119–144.

Arthur, W. B. (1989). Competing technologies, increasing returns, and lock-

in by historical events. The economic journal 99 (394), 116–131.

Arthur, W. B. (1994). Increrasing returns and path dependence in the econ-

omy. University of Michigan Press.

Banerjee, A. and E. Duflo (2011). Poor economics: A radical rethinking of

the way to fight global poverty. PublicAffairs Store.

32

Bates, R. H. (2001). Prosperity and violence: the political economy of devel-

opment. WW Norton New York.

Besley, T. and T. Persson (2010). State capacity, conflict, and development.

Econometrica 78 (1), 1–34.

Besley, T. and T. Persson (2011). Pillars of prosperity: The political eco-

nomics of development clusters. Princeton University Press.

Bourdieu, P. and R. Nice (2000). Acts of resistance: Against the new myths

of our time. Taylor & Francis.

D’Agnese, L. and R. Abravanel (2011). Regole. Garzanti.

Dardi, M. and F. Targetti (1984). Domanda. Boringhieri.

David, P. A. (1985). Clio and the economics of qwerty. The American

economic review 75 (2), 332–337.

David, P. A. (1994). Why are institutions the carriers of history?: Path de-

pendence and the evolution of conventions, organizations and institutions.

Structural change and economic dynamics 5 (2), 205–220.

Dei, F. (2011). Pop-politica: le basi culturali del berlusconismo. Studi cul-

turali 8 (3), 471–490.

Diamond, J. (2005). Collapse: how societies choose to fail or succeed: revised

edition. Penguin. com.

Dixit, A. and B. Nalebuff (2008). The art of strategy: A game theoristss

guide to success in business and life. New York: Norton.

33

Dubois, D., D. D. Rucker, and A. D. Galinsky (2012). Super size me: product

size as a signal of status. Journal of Consumer Research 38 (6), 1047–1062.

Festinger, L. (1962). A theory of cognitive dissonance, Volume 2. Stanford

university press.

Fiorito, L. and M. Vatiero (2013). A joint reading of positional and rela-

tional goods. In Economia politica: Journal of analytical and institutional

economics.

Gambetta, D. and G. Origgi (2013). The ll game the curious preference for

low quality and its norms. Politics, Philosophy & Economics 12 (1), 3–23.

Ginsborg, P. (1998). L’Italia del tempo presente: famiglia, societa civile,

Stato, 1980-1996, Volume 499. Einaudi (IT).

Haavelmo, T. (1950). The notion of involuntary economic decisions. Econo-

metrica: Journal of the Econometric Society 18, 1–8.

Hirschman, A. O. (1967). The principle of the hiding hand. Brookings Insti-

tution.

ISTAT (2013). Il benessere equo e sostenibile in italia, rapporto 2013. Tech-

nical report.

Janis, I. L. and L. Mann (1977). Decision making: A psychological analysis

of conflict, choice, and commitment. Free Press.

Kilgour, D. M. and N. M. Fraser (1988). A taxonomy of all ordinal 2× 2

games. Theory and decision 24 (2), 99–117.

34

Lave, C. A. (1985). Speeding, coordination, and the 55 mph limit. The

American Economic Review 75 (5), 1159–1164.

Leibenstein, H. (1950). Bandwagon, snob, and veblen effects in the theory of

consumers’ demand. The Quarterly Journal of Economics 64 (2), 183–207.

Lewin, K. (1946). Action research and minority problems. Journal of social

issues 2 (4), 34–46.

Marshall, T. H. (1964). Class, citizenship and social development. Chicago:

University of Chicago Press.

Martinko, M. J. and W. L. Gardner (1982). Learned helplessness: An al-

ternative explanation for performance deficits. Academy of Management

Review 7 (2), 195–204.

North, D. C., J. J. Wallis, and B. R. Weingast (2009). Violence and social

orders: a conceptual framework for interpreting recorded human history.

Cambridge University Press.

Olson, M. (1965). The logic of collective action. Public goods and the theory

of groups. Harvard University Press.

Pagano, U. (1999). The politics and economics of power, Chapter Is power an

economic good? Notes on social scarcity and the economics of positional

goods. Routledge.

Patterson, O. (2010). The mechanisms of cultural reproduction. explain-

ing the puzzle of persistence in hall, grindstaff and lo (eds), handbook of

cultural sociology.

35

Pianta, M. (2012). Nove su dieci. Perche stiamo (quasi) tutti peggio di dieci

anni fa. Bari, Laterza.

Pierson, P. (2000). Increasing returns, path dependence, and the study of

politics. American political science review , 251–267.

Putnam, R. D., R. Leonardi, and R. Y. Nanetti (1994). Making democracy

work: Civic traditions in modern Italy. Princeton university press.

Rapoport, A. (1988). Decision theory and decision behaviour. Macmillan,

London.

Rapoport, A. and M. Guyer (1967). A taxonomy of 2 x 2 games. Bobbs-

Merrill.

Robinson, D. and D. Goforth (2005). The Topology of the 2× 2 Games. New

York: Routledge.

Rosen, S. (1981). The economics of superstars. The American economic

review 71 (5), 845–858.

Rucker, D. D. and A. D. Galinsky (2008). Desire to acquire: Powerlessness

and compensatory consumption. Journal of Consumer Research 35 (2),

257–267.

Tilly, C. (1975). Reflections on the history of european state-making. The

formation of national states in Western Europe, 3–83.

Uhlaner, C. J. (1989). Relational goods and participation: Incorporating

sociability into a theory of rational action. Public choice 62 (3), 253–285.

36